Comparing environmental impacts: in situ thermal desorption, in situ chemical oxidation/reduction, and ex situ desorption for chlorinated hydrocarbon-contaminated site

Abstract

The integration of on-site data with optimized remediation activities provides technical and administrative support for advancing green and sustainable remediation (GSR). This study quantified the environmental impacts of remediating a chlorinated hydrocarbon-contaminated site using three technologies: in-situ chemical oxidation/reduction (ICOR), in-situ thermal desorption (ISTD), and ex-situ desorption (ED), assessed with the Spreadsheet for Environmental Footprint Assessment (SEFA) and SiteWise™ tools. The project generated 100.02 t of air pollutant, 15,936.29 t CO₂ eq of greenhouse gas (GHG) emissions, and required 8.085 × 10⁷ MJ of energy and 1.1 × 106 L of water. Energy use and GHG emissions were highest for ISTD and lowest for ED, with ICOR showing intermediate values. Per cubic meter of treated soil, energy use and GHG emissions were 607.06 MJ/m³ and 138.80 kg CO₂ eq/m³ for ISTD, 289.40 MJ/m³ and 31.35 kg CO₂ eq/m³ for ED, and 315.28 MJ/m³ and 42.44 kg CO₂ eq/m³ for ICOR. When normalized per unit pollutant removed, ISTD showed higher efficiency than ED and ICOR. Major contributors to energy use and GHG emissions included heating process in ISTD, barriers materials in ICOR, and remediation materials in ED and ICOR. Optimization measures include adding lateral insulation, adopting precision and moderate remediation schemes, and using renewable energy.

Introduction

Long-term industrial activities have resulted in extensive land contamination worldwide^1,2. Various techniques—such as soil washing, chemical oxidation/reduction, stabilization/solidification, thermal remediation, electrokinetic remediation, and bioremediation—have been proven as effective soil remediation strategies^3–5. However, the selection of a particular remediation technology is contingent upon the specific characteristics of each contaminated site⁶.

When undertaking soil remediation projects, it is essential not only to evaluate the effectiveness of the selected techniques but also to carefully assess their potential adverse environmental impacts. Indeed, some studies have indicated that these secondary impacts may surpass the main beneficial effects of the selected technology^7–9. The recently introduced concept of green and sustainable remediation (GSR) The recently introduced concept of GSR emphasizes the need to minimize the secondary impacts of remediation activities—for example, by incorporating key sustainability indicators such as energy efficiency, air and water resource conservation, material and waste reduction, land and ecosystem protection, greenhouse gas (GHG) emissions reduction, cost-effectiveness, and considerations of human health, safety, and social equity, and has attracted increasing attention from both governmental agencies and industries^10–12. GSR refers to a set of principles and practices aimed at optimizing the use of natural resources and energy while minimizing negative environmental impacts, all without compromising remediation effectiveness^13–15.

Environmental impact assessments are pivotal for enhancing the sustainability of remediation efforts^2,12,16. Among the key instruments available for such evaluations are the Sustainable Environmental Footprint Assessment (SEFA) and the Excel-based tool SiteWise™^14,17,18. Developed by the United States Environmental Protection Agency (USEPA), SEFA comprises a series of spreadsheets designed to quantify the environmental impact associated with contaminated site remediation¹⁹. In parallel, SiteWise™, developed jointly by the US Navy, the US Army Corps of Engineers, and the applied science and technology organization Battelle Memorial Institute, a private non-profit R&D organization in the United States, calculates the environmental impact of remedial alternatives commonly employed in the remediation industry²⁰. Both SEFA and SiteWise™ integrate the evaluation of resource consumption with the quantification of pollutant emissions, which are considered as secondary impacts, across the full life cycle of a specific remediation activity^14,21. Nonetheless, retrospective evaluations employing SEFA or SiteWise™ to analyze the environmental impacts of completed contaminated site remediation projects remain notably scarce. Moving forward, it would be insightful to examine how these tools perform across different remediation scenarios and investigate opportunities for methodological enhancements via further integration of sustainability metrics.

It is well established in the literature that soil remediation activities have a significant environmental impact^6,22,23. Energy consumption and GHG emissions during soil remediation vary depending on the remediation process and the types of pollutants present. In particular, certain pollutants are more recalcitrant, requiring extra remediation steps or materials^6,24. Inoue and Katayama²⁵ evaluated four remediation techniques, i.e., disposal, thermal desorption, biopiles, and landfarming, in terms of energy consumption and CO₂ emissions derived from them and found that biopile-based technology was the most energy-efficient and generated the lowest CO₂ emissions. Mauko et al.²⁶ compared a proposed in situ remediation scenario with two alternative reclamation scenarios and a no-action option and demonstrated that the in situ remediation scenario had significantly lower impacts on human health and ecosystem quality. Another study by Khan et al.¹⁴ analyzed the integration of renewable energy sources into remediation system design, showing that this approach significantly reduced environmental impacts. Multiple technologies are often needed to address varying levels of pollution. In practice, the environmental impacts of remediation projects are typically assessed by comparing the performance of the different technologies employed^7,12,15,27. In this study, we employed SEFA and SiteWiseA™ as analytical tools to evaluate the environmental impact of a contaminated site remediation project conducted in Tianjin city during 2019. The main objectives were: (i) to determine the overall environmental impact of the remediation technologies in-situ thermal desorption (ISTD), in-situ chemical oxidation/reduction (ICOR) and ex-situ desorption at room temperature (ED); (ii) to compare the environmental impacts of three remediation technologies, expressed per cubic meter of contaminated soil treated and per kilogram of contaminant removed. (iii) to examine the underlying causes of and main contributors to the impacts of these remediation alternatives; and (iv) to assess the value of evaluating impacts before and after remediation while proposing methods to optimize the process. Given the large number of contaminated sites in Tianjin requiring remediation, yet lacking empirical case studies on environmental footprint analysis, the conclusions of this study will provide a valuable reference for the green and sustainable remediation (GSR) of contaminated sites in Tianjin.

Materials and methods

Site description

The contaminated site subjected to remediation is located in an abandoned chemical factory in Xiqing District, Tianjin, China, and has an area of 34,100 m². Owing to its marine sedimentary structure, this coastal site is characterized by complex low-permeability strata and a prevailing high groundwater level. From 1956 to 2011, this facility produced over 350 types of fine organic chemical products, resulting in significant contamination of the surrounding soil and groundwater with volatile organic compounds (VOCs), particularly chlorinated hydrocarbons. Specifically, 9 and 31 VOCs were detected in the soil and groundwater, respectively. The pollutant concentrations and remediation target values are detailed in Tables S1 and S2. Permission to conduct sampling and experimentation at this contaminated site was secured from the relevant authorities.

Remediation alternatives

Due to pollution levels and geological structures varying with soil depth at the study site, we selected three targeted remediation strategies: ED for 7,579.1 m³ of soil, ISTD for 21,812.2 m³ of soil and 63,125.3 m³ of aquifer; and ICOR for 73,797.9 m³ (oxidation) and 17,135.4 m³ (reduction) of aquifer. Further details on these alternatives and the specific volumes of treated soil are provided in Table 1 and Fig. S1.

Table 1.

Summary of the remediation technologies selected for the contaminated site.

Remediation alternatives	Remediation method	Contaminated unit	Remediation volume (m3)	Quantity of pollutants removed (kg)
1	Ex-situ desorption at room temperature	0–2 m contaminated soil	7579.1	303.16
2	In-situ thermal desorption	2–14 m contaminated soil, 3–13 m contaminated groundwater and 3–7 m Non-aqueous phase liquid (NAPL) area	84937.4	41750.52
3	In-situ chemical oxidation/reduction	3–7 m contaminated groundwater	90933.3	4459.55

ED was employed for remediating soil in the 0–2 m layer, with all operations conducted in an enclosed on-site facility to prevent fugitive emissions. During remediation, a 7% lime desorption agent was mixed with the contaminated soil and agitated mechanically, promoting the volatilization of VOCs adsorbed on the soil particles at room temperature. Several studies have demonstrated that lime is an effective soil amendment for removing organic contaminants^28–30. The exhaust gas containing the volatilized pollutants was then collected using a ventilation pipeline and treated with granular activated carbon (GAC)^31,32.

ICOR was used to treat lightly and moderately contaminated aquifers. Specifically, chemical oxidation was adopted for lightly contaminated aquifers using sodium persulfate, an oxidizing agent shown to be effective against VOCs^17,33, at a dosage of 0.3% to 0.5% by weight, with adjustments made to suit the site conditions, whereas chemical reduction was employed to treat moderately contaminated aquifers using a composite reductant made of 30% zero-valent iron (ZVI, Fe⁰) and 70% yeast^34,35. A high-pressure rotary jetting pile was used to inject the oxidant or reductant into the polluted soil and aquifer, with injection holes spaced 3.5 m apart (as determined by pilot tests). The redox agents then diffused in situ and reacted with the pollutants, breaking them down into less toxic or non-toxic products such as CO₂, water, or chloride ions^36–38.

ISTD was used to treat contaminated soil at depths of 2–14 m, severely polluted aquifers (with detected non-aqueous phase liquids), and moderately polluted aquifers affected by soil contamination. Heating wells equipped with heating rods were installed in the contaminated area, and the site was heated to 90–110℃ to promote the decomposition and volatilization of pollutants from soil and aquifers into the gas phase³⁹. The volatile substances were then extracted at flow rates between 5000 and 10,000 m³/h and purified using equipment for exhaust gas treatment that combines combustion and GAC^31,32.

Assessment approach

A retrospective Life Cycle Assessment (LCA) of a full-scale contaminated site remediation project was conducted based on SURF^19,40 and SEFA^14,20. The United States Sustainable Remediation Forum (US SURF) provides a framework that integrates environmental, social, and economic considerations into site cleanup. The assessment of the environmental impact of different remediation technologies follows seven key steps: (i) define goals and scope, specifying remediation objectives, regulatory requirements, and sustainability targets; (ii) establish system boundaries, delimiting spatial and temporal coverage, site conditions, remediation phases, and life-cycle stages; (iii) identify project metrics, including environmental (GHG emissions, water use), economic (cost-effectiveness), and social (community impact) indicators; (iv) compile the project inventory, gathering site, contaminant, technology, and resource input data; (v) quantify inputs/outputs using SEFA or SiteWise; (vi) assess impacts via LCA for air, water, soil, ecosystems, and human health; and (vii) analyze results to compare alternatives and identify optimization opportunities. Specifically, the diagram in Figs. 1 and 2 illustrates the site boundaries and the flows of environmental inputs (energy and resources) and outputs (emissions, waste) for the three remediation technologies examined (i.e., ED, ICOR, and ISTD) and temporary construction (TC, comprising temporary access roads, administrative offices, and workshop foundations, etc.). The system boundaries and project metrics differed between the overall project assessment and the assessment of individual remediation technologies. The visual depiction in Fig. 2 aids in understanding the environmental inputs and outputs associated with each remediation method.

Fig. 1.

Aerial view of the contaminated site subjected to remediation in Xiqing District, Tianjin. The base satellite imagery was generated using Google Earth (web version 10.102.78.2, Google LLC; https://earth.google.com/web/). Date of imagery acquisition: January 12, 2026. Annotations were added by the authors using Autodesk AutoCAD 2020 (Autodesk, Inc.; https://www.autodesk.com).

Fig. 2.

Schematic diagram illustrating the system boundaries of the LCA for three remediation technologies, and the calculation components used to assess their environmental impacts.

To quantify the sensitivity of various factors contributing to the environmental footprint, a sensitivity analysis was performed using Eq. (1).

1

where SR is the parameter sensitivity (dimensionless); P₁ and P₂ represent the parameter values before and after the variation, respectively; and Y₁ and Y₂ are the quantified environmental footprint results calculated based on P₁ and P₂.

Data analysis

Data were collected during each remediation process and classified into four categories: material production, transportation, equipment operation, and other processes. In order to compare the selected remediation technologies without bias due to the volume of treated soil, energy consumption was normalized to a cubic meter of contaminated soil²⁷. The material production category included data on infrastructure, well construction materials, equipment, and remediation materials. For this category, manufacturing and depreciation factors were considered for special equipment only and not for commonly used equipment. Transportation data covered the movement of personnel, equipment, and materials, with parameters selected from the SiteWise^™ list. Equipment operation data concerned the use of excavation and transfer machinery, remediation equipment, and gas treatment systems. Energy consumption for equipment operation was recorded based on actual usage, and parameters for common equipment (e.g., excavators and dozers) were obtained from the SiteWise^™ list. The electricity-generation mix was calculated to reflect local conditions. Finally, the data for “other processes”, for example related to waste treatment and sample analysis, were calculated according to SEFA guidelines. Emission and energy consumption coefficients were selected from either the SiteWise^™ list (preferably) or the SEFA list. A summary of the entire remediation project, including all applied technologies, is provided in Table 2, while Table S3 lists detailed data on energy consumption (fuel and electricity) and resource inputs such as materials, water, and equipment.

Table 2.

Summary of data related to the site remediation project including different technologies.

	Operation/Material	Process included in LCI	Amount	Unit
Process 1 Temporary construction (TC)	Materials	Infrastructure materials	203.925	t
		Water	100	t
	Transportation	Transportation of personnel and materials
	Equipment operation	Dozer, loader and crawler crane
Process 2 Ex-situ desorption at room temperature (ED)	Materials	Infrastructure materials	206.003	t
		Equipment materials	20	t
		Remediation materials	986	t
	Transportation	Transportation of personnel, equipment and materials
	Equipment operation	Infrastructure process: dozer, loader, crawler crane and welder (8 kw)
		Soil excavation and refill process: dozer and loader
		In-site transportation of soil: 7579.1 m3, 500 m
		Soil remediation and gas treatment
	Other processes	Waste treatment (abandoned GAC)	5	t
		Analysis and detection process	35	samples
Process 3 In-situ chemical oxidation/reduction (ICOR)	Materials	Remediation materials	260	t
		Well materials	290	kg
		Vertical barrier materials	4080	t
		Water	1000	t
	Transportation	Transportation of personnel, equipment and materials
	Equipment operation	Vertical barrier construction process: 340 kw three-axis stirring pile
		Well construction process
		Soil stirring and remediation process: high pressure rotary jet pile and blender
	Other processes	Analysis and detection process	100	samples
Process 4 In-situ thermal desorption (ISTD)	Materials	Infrastructure materials	17.24	t
		Insulation materials	5142.91	t
		Vertical barrier materials	787.5	t
		Well materials	481.71	t
		Equipment materials	21.74	t
		Remediation materials	10	t
	Transportation	Transportation of personnel, equipment and materials
	Equipment operation	Vertical barrier construction process: 340 kw pile mixer
		Well construction process
		In-situ thermal treatment: 12 kw electric heating rods
		Exhaust gas treatment
	Other processes	Waste treatment (waste GAC)	40	t
		Sample analysis	220	samples

Results and discussion

Environmental life-cycle comparisons of integral remediation project

The total amount of energy consumed during the remediation project was 8.085 × 10⁷ MJ, with ISTD, ICOR, ED accounting for 63.8%, 32.6%, 3.0%, and temporary construction (TC) accounts for 0.7% (Fig. 3a). Energy consumption during each remediation process depended on the materials used, equipment employed, and volume of treated soil. For example, only 7579.1 m³ of soil was treated during ED, whereas 84937.4 m³ and 90933.3 m³were treated during ISTD and ICOR, respectively, which explains the considerably lower energy use during the former process. Fuel and electricity accounted for 52% and 48% of the total energy consumption, respectively. Notably, 98.68% of electricity was used for operating the ISTD system. Energy consumption normalized to a cubic meter of contaminated soil was highest during ISTD (607.06 MJ/m³), whereas lower and comparable values were observed during ED (289.40 MJ/m³) and ICOR (315.28 MJ/m³) (Fig. 3b). Although the volumes of treated soil during ISTD and ICOR were similar, the total amount of energy consumed during the former process was nearly twice as high. Interestingly, energy consumption per unit mass of pollutant removed was much lower during ISTD than during ED and ICOR (Fig. 3c). This suggests that, although ISTD is energy-intensive, it is effective at removing organic pollutants. Overall, in order to optimize the efficiency of remediation, it is necessary to consider both the energy consumption and pollutant removal efficiency of the selected technology.

Fig. 3.

Energy consumption metrics by remediation technology: (a) total energy, (b) energy per unit volume of treated soil, and (c) energy per unit mass of pollutant removed.

Overall, the remediation project generated a total of 15,936 t of GHG emissions. ISTD accounted for the largest proportion of emissions, followed ICOR, ED, which corresponds to the energy consumption pattern of these technologies (Fig. 4a). Specifically, ISTD and ICOR were the primary contributors, accounting for 73.98% and 24.22% of the total GHG emissions, respectively (Fig. 4a). Independent of the volume of contaminated soil, the GHG emissions produced per unit volume of treated soil were 138.80 kg CO₂ eq/m³ for ISTD, 31.35 kg CO₂ eq/m³ for ICOR, and 42.44 kg CO₂ eq/m³ for ED (Fig. 4b). Although in situ remediation is typically considered as a less disruptive approach with a lower environmental impact^6,41, our evaluation indicates that ISTD and ICOR produced higher GHG emissions per unit volume of soil compared to ED. ED technology under normal temperature is designed to treat lightly contaminated soil, which requires only a small amount of applied chemicals (68.68 t of lime). In contrast, the treatment of heavily contaminated soil, for example via ISTD and ICOR would demand a significantly higher chemical consumption to achieve similar remediation results. However, the GHG emissions per unit mass of pollutant removed generated by ISTD were significantly lower than those produced by ED and ICOR (Fig. 4c), this indicates that thermal desorption is the most environmentally sustainable remediation technology among those evaluated for severely contaminated soil, underscoring the importance of assessing environmental impacts using multiple performance metrics.

Fig. 4.

GHG emissions by remediation technology: (a) total GHG emissions, (b) GHG emissions per unit volume of treated soil, and (c) GHG emissions per unit mass of pollutant removed.

An emission inventory provides essential data for developing effective emission reduction strategies. Table 3 presents the levels of emission of critical air pollutants, i.e., nitrogen oxides (NO_x), sulfur oxides (SO_x), and particulate matter (PM)²⁰, and water consumption associated with the remediation project. Our results indicate that NO_x, SO_x, and PM emissions were considerably higher during ISTD and ICOR. In particular, the ISTD process accounted for 80.68% and 83.99% of NO_x and SO_x emissions, respectively. This can be attributed to the use of fossil fuels—namely coal, natural gas, and oil—for electricity generation^42,43. PM emissions primarily derive from the manufacturing of materials for insulation, vertical barrier construction, and well construction⁴³. During ICOR, NO_x, SO_x, and PM emissions were primarily due to the production of materials for vertical barrier construction (specifically, cement production), accounting for over 90% of the total emissions⁴⁴. In terms of water consumption, 1 × 10⁵ L was required for preparing the oxidizing/reducing agents for ICOR. Since the water in this project was directly injected into the ground, there was no need for wastewater recovery and treatment.

Table 3.

Emissions of critical air pollutants and water consumption for each tested remediation technology.

Parameter	TC	ED	ICOR	ISTD
Total NOx emission (kg)	172.61	634.27	7484.37	21178.96
Total SOx emission (kg)	149.46	732.22	14989.84	49808.73
Total PM emission (kg)	37.56	137.68	2951.69	1741.65
Unit NOx emission (kg/m3)	--	0.08	0.08	0.25
Unit SOx emission (kg/m3)	--	0.10	0.16	0.59
Unit PM emission (kg/m3)	--	0.02	0.03	0.02
Water consumption (t)	100	0	1000	0

Environmental footprint of different remediation stages

We assessed the environmental impacts of the three remediation technologies (Fig. 5), each defined by distinct system boundaries (Fig. 1). These analyses considered the full life cycle (cradle-to-grave) of all activities within each remediation process²⁷.

Fig. 5.

Sankey diagram illustrating the environmental impacts of ED, ICOR, and ISTD in terms of energy consumption and GHG emissions (a–f).

Energy consumption and GHG emissions were primarily associated with material production and equipment operation. Specifically, with the production of 981 t of 7% lime⁴³ and both soil excavation and refilling. These activities consumed 9.37 × 10⁵ MJ (123.66 MJ/m³) and 6.6 × 10⁵ MJ (87.61 MJ/m³) of energy, accounting for 39.21% and 27.78% of the total energy use, respectively. The corresponding GHG emissions were 86,870.12 kg CO₂ Eq. (11.46 kg CO₂ eq/m³) and 47,338.70 kg CO₂ Eq. (6.25 kg CO₂ eq/m³), accounting for 36.56% and 19.92% of the total emissions, respectively. Soil remediation and gas treatment processes accounted for 22.61% of the total GHG emissions. Of these, 76.07% originated from electricity use for dust removal and exhaust gas treatment equipment, and the remaining 23.93% derived from fuel use in soil mixing operations. In contrast, the transportation process during ED contributed less than 3% (7.74 MJ/m³ and 0.05 kg CO₂ eq/m³), due to the adoption of an on-site approach. This aligns with the findings of Vocciante et al.²⁷ reporting that transportation can significantly affect remediation emissions (39.9 kg CO₂ eq/m³) and that this issue can be minimized through on-site implementation.

Material production—particularly for vertical barrier construction—was the main contributor to environmental impact, accounting for 71.32% (1.88 × 10⁷ MJ, 206.39 MJ/m³) of energy use and 87.75% (3.39 × 10⁶ kg CO₂ eq, 37.24 kg CO₂ eq/m³) of emissions. A vertical barrier was constructed along the site boundary to control plume migration and was included in the environmental assessment of ICOR. Since the treatment area for this remediation technology was considerably larger than that for other technologies and closely matched the site boundary (Fig. 1), vertical barrier construction was considered as a significant influencing factor. Excluding the vertical barrier, the soil treatment process alone consumed 78.74 MJ/m³ of energy and resulted in 5.08 kg CO₂ eq/m³ of GHG emissions, primarily derived from the production of materials for remediation.

Figure 5 (e and f) shows the environmental impacts of all activities during the ISTD process. According to the data, the heating equipment used for thermal treatment was the principal contributor, accounting for 72.38% of the total energy consumption (3.73 × 10⁷ MJ; 439.40 MJ/m³) and 82.84% of the total GHG emissions (9.76 × 10⁶ kg CO₂ eq; 114.97 kg CO₂ eq/m³). The corresponding values per unit volume of treated soil were 439.40 MJ/m³ and 114.97 kg CO₂ eq/m³, respectively. Amponsah et al.⁶ reported a maximum level of GHG emissions of 200 kg CO₂-eq/m³ during in situ thermal remediation. In the present study, a lower value was obtained (138.80 kg CO₂ eq/m³), likely reflecting differences in the energy mix employed. Several studies have shown that GHG emissions from in situ thermal remediation can substantially increase depending on the electricity and heat sources used^5,6,45. Moreover, variations in emission levels can be attributed to factors such as the type and concentration of pollutants, heating temperature, and duration of the remediation process. However, in terms of pollutant removal efficiency, energy consumption and GHG emissions per unit mass of pollutant removed during ISTD were relatively low, indicating an acceptable performance. Fig. 6

Fig. 6.

Sensitivity analysis of environmental footprint indicators for each remediation technology.

A sensitivity analysis was performed to evaluate the environmental footprint contributions of the three remediation technologies. The results indicate that ISTD exhibits the highest sensitivity regarding GHG, NOx, and SOx emissions, as well as energy consumption, followed by ICOR and ED. In contrast, for PM10 emissions, ICOR demonstrates the greatest sensitivity, surpassing both ISTD and ED. Overall, the findings suggest that mitigating the environmental footprint of ISTD is the most effective strategy for reducing the total environmental burden of this project.

Optimization of remediation processes based on environmental footprint assessment

Environmental impact assessment plays an important role in the optimization of remediation processes, as it promotes energy conservation, emission reduction, and the advancement of GSR practices. Optimization can be categorized into two types depending on the targeted stage of remediation: prospective (during the pre-construction stage) or retrospective (during the post-construction stage). Prior to project implementation, a prospective assessment based on estimated engineering data, including energy consumption and GHG emissions, can be conducted to evaluate the environmental impact. Based on the data obtained, key drivers of energy use and emissions are identified, allowing the development of targeted optimization strategies. However, this method involves inherent uncertainties due to the reliance on estimated parameters. In contrast, once the remediation project is completed, a retrospective assessment using actual engineering data (related for example to material inputs and energy use) enables a more accurate analysis of the main contributors to energy consumption and GHG emissions. This, in turn, provides valuable insights for improving future projects and supports the refinement of prospective evaluation methodologies. The overall uncertainty of retrospective optimization is minimal, primarily arising from secondary parameters such as estimated transportation distances, equipment operational profiles, and other non-critical inputs.

In this study, a retrospective environmental impact assessment was performed upon completion of specific remediation activities. The results revealed the primary drivers of energy consumption and GHG emissions (Fig. 7). The most significant contributor to both total energy consumption and GHG emissions was the heating process during ISTD. Additional significant emissions derived from the production of materials for vertical barrier construction during ICOR. The manufacturing of remediation materials during both ED and ICOR also contributed substantially to the environmental impact. Recent advances in GSR provide a comprehensive, technology-driven strategic plan for the recovery of contaminated sites. Key directions to further minimize the environmental impact of combined ISTD, ICOR, and ED schemes are outlined below.

Fig. 7.

Flowchart of the optimization of a remediation process based on environmental impact assessment.

i)Maximizing thermal efficiency during ISTD. A combination of strategic scheduling, advanced control, and targeted insulation can be employed to this end. Lemming et al.²³ have demonstrated that scheduling heating cycles outside periods of peak electricity demand, deploying vapor caps with reduced concrete content can lower the overall environmental impact and resource depletion by 10–20%, respectively. Furthermore, optimizing the spacing of heating wells as well as operation schedules—coupled with condensate capture and reuse in thermal conduction or steam-generation loops—minimizes baseline energy demand. Real-time thermal mapping via distributed fiber-optic sensing enables dynamic adjustment of power to individual heater strings only where subsurface temperatures are below target temperature, yielding an additional 15–20% reduction in energy input. Finally, surrounding heating wells with advanced lateral insulation (vacuum-insulated panels or aerogel blankets) can cut conductive heat loss by up to 30%^23,46.

• ii)Precision chemistry to mitigate the material-related environmental impact during ED and ICOR. Passive samplers and high-resolution mass spectrometry can be used to produce contaminant concentration maps at sub-meter resolution. Injection volumes can then be directly calibrated against these fine-scale concentration maps to minimize total chemical input and prevent over-remediation. Furthermore, it is possible to conduct dose titration experiments and apply reactive-transport models in order to select optimal reagents and injection strategies, which can significantly reduce chemical usage by 25–35%^47–49. Finally, biodegradable surfactants or redox-active biopolymers should be adopted as alternatives to synthetic oxidants, thereby lowering secondary waste generation and toxicity risks.
• iii)Integrating on-site renewable energy with hybrid power systems can markedly reduce the carbon footprint of thermal remediation technologies. As outlined in the USEPA’s Green Remediation Best Management Practices, co-locating solar photovoltaic arrays and small-scale wind turbines provides off-grid electricity for heating, pumping, and monitoring. USEPA also recommends quantifying a project’s environmental impact and minimizing it by adopting renewable energy technologies. (e.g., solar photovoltaics for auxiliary power). Empirical studies have confirmed that shifting the energy mix, especially by incorporating solar photovoltaics, significantly lowers both overall Primary energy demand and GHG emissions^{14,42,50–52}.
• iv)Coupling thermal, chemical, and microbial processes for sustainable soil remediation. Coupling ISTD with chemical oxidation and microbial processes offers a robust method for sustainable remediation of contaminated sites.Lemming et al.²³ revealed that optimized ISTD–ISCO sequences lowered CO₂-equivalent emissions by 20% and resource use by 15% compared to excavation or long-term vapor extraction. By strategically alternating thermal, chemical, and biological stages, practitioners can maximize mass removal rates, minimize secondary impacts, and advance toward carbon-neutral operations.

Conclusions

In this study, we conducted a rigorous retrospective LCA of a full-scale contaminated site remediation project employing various remediation techniques. The assessment utilized site-specific, field-collected data and was performed using the SEFA and SiteWise™ tools. The project’s total GHG emissions and energy consumption were 15936.29 t CO₂ eq and 8.08 × 10⁷ MJ, respectively. This technology resulted in the highest energy consumption per cubic meter of treated soil (607.06 MJ/m³), followed by ED (315.28 MJ/m³) and ICOR (289.40 MJ/m³). The corresponding GHG emissions per unit volume of treated soil were 0.14 kg CO₂ eq/m³ for ISTD, 31.35 kg CO₂ eq/m³ for ED, and 42.44 kg CO₂ eq/m³ for ICOR. Sensitivity analysis revealed that ISTD contributed the most to the overall environmental impact, followed by ICOR and ED. In terms of data normalized per unit mass of pollutant removed, ISTD emerged as the most environmentally sustainable technology. From the perspective of unit pollutant removal efficiency, in-situ thermal desorption (ISTD) proves to be the most sustainable methodology for the remediation of severely contaminated soil (source zones). From an engineering standpoint, the main drivers of energy consumption and GHG emissions were the heating process during ISTD, the production of materials for vertical barrier construction in ICOR, and the manufacturing of remediation agents in both ED and ICOR. To optimize future remediation projects, we recommend (1) constructing lateral insulation layers for ISTD, (2) implementing precision-based and moderate remediation schemes, and (3) integrating renewable energy sources. These strategies will also contribute to advancing green and sustainable practices for contaminated site remediation. Since this study focused on the physical and chemical mechanisms of ISTD, ICOR, and ED, microbial succession was not monitored. Future research incorporating biological indicators will provide a more comprehensive assessment of the long-term ecological footprint.

Author contributions

Peng Liu: Conceptualization, methodology, data curation, writing - original draft, writing - review & editing. Xin Li: Data curation. Meng Xiao: Data curation. Xintong Yang: Methodology. Hongzhen Zhang: Writing - review & editing. Xianglan Li: Writing - review & editing.

Funding

The research is supported by The National Key Research and Development Program of China (Nos. 2022YFC3703300).

Data availability

The data analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Supplementary Information

Supplementary Information to this article can be found in file supplementary material.